Workshop on Mathematics for Petascale Data, June 3-5, 2008, Rockville, MD

这位研究人员和她的同事提议组织一次研讨会，以确定未来数学进步的机会，这是从千万亿级科学数据中进行科学发现所必需的。研讨会将包括来自学术界、国家实验室和工业界的科学家。应用科学家、数学家、统计学家、高性能数据挖掘者将确定一个研究议程，以开发应对千万亿级数据集带来的挑战所需的下一代数学技术。他们将编写一份研讨会报告，该报告将：*阐明各种科学领域的要求，如全球碳循环建模、聚变能源生产、国家安全等。*描绘适当的数学方法和技术，*确定这些方法和技术的当前最新水平，并*确定必须解决的差距，以便在未来五到十年内对大型、复杂的数据集进行有效分析。可再生能源生产、碳封存、国家安全和人类健康保护是当今社会的紧迫问题。主要诱因--经济、地缘政治、和环境--推动研究和开发具有成本效益和效益的解决方案的科学任务。必须真正综合科学发现的三大支柱--实验、理论和超大规模计算--以有效和全面地应对这些挑战。每一支柱的进步都已经彻底改变了科学的开展方式。然而，伴随着这一承诺而来的是一个问题--国家高通量实验学院、天文台和UltraScale计算设施如此产生的海量数据。这些数据提供了有关宇宙本质的基本问题的答案。然而，答案将微妙地隐藏在原始数据中。这些数据需要被分析以提取知识-理解科学。要发现这样的新知识将需要来自几个领域的下一代数学技术，包括但不限于统计学、机器学习、图像分析和模式识别。这些领域的进展还将使用实验和观测数据反复验证超尺度模拟成为可能。因此，这些技术将为我们在健康、能源、环境和国家安全方面面临的一些最紧迫和最昂贵的挑战带来革命性和非常规的解决方案。

The investigator and her colleagues propose to organize a workshop to identify opportunities for future mathematical advances required for enabling scientific discoveries from petascale scientific data. The workshop will embrace scientists from academia, national laboratories, and industry. Application scientists, mathematicians, statisticians, high performance data miners will define a research agenda for developing the next-generation mathematical techniques needed to meet the challenges posed by petascale data sets. They will produce a workshop report that will:* Articulate the requirements of various scientific domains suchas global carbon cycle modeling, fusion energy production, nationally security, etc.* Delineate appropriate mathematical approaches and techniques,* Determine the current state-of-the-art in these approaches andtechniques, and* Identify the gaps that must be addressed to enable the effectiveanalysis of large, complex data sets in the next five to ten years.Renewable energy production, carbon sequestration, national security, and human health protection are the urgent issues in today's society.Major incentives-economic, geopolitical, and environmental-drive a scientific mandate to research and develop cost-effective and beneficial solutions. It is imperative to truly synthesize the three pillars of scientific discovery-experimentation, theory, and ultra-scale computation-to address these challenges effectively and comprehensively.Advances in each have been already revolutionizing the way science is conducted. With this promise, however, comes a problem - the massive quantities of data so produced by national high-throughput experimental faculties, observatories and ultrascale computing facilities. Those data hold the answers to fundamental questions about the nature of the universe. However, the answers will be subtly hidden in the raw data.Those data need to be analyzed to extract knowledge - to understand the science. Discovering such new knowledge will require the next-generation mathematical techniques from several fields, including but not restricted to statistics, machine learning, image analysis, and pattern recognition. Advances in these areas will also enable iteratively validating ultra-scale simulations with experimental and observational data. As a result, these technologies will bring revolutionary and unconventional solutions to some of our most pressing and expensive challenges in health, energy, environment, and national security.